ModelRefs / Prompt Injection — AI Glossary
Prompt Injection — AI Glossary
An attack where untrusted content in retrieved documents or tool outputs overrides or hijacks the model's instructions. OWASP's #1 LLM security risk.
Overview
OWASP's #1 LLM security risk. Direct injection overwrites the system prompt; indirect injection embeds instructions in retrieved web pages or documents. Defenses: clear delimiters, instruction hierarchies, never trusting tool output as instructions.
Reference details
| Topic | safety |
|---|---|
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Prompt Injection — AI Glossary.
Frequently asked questions
What is Prompt Injection?
An attack where untrusted content in retrieved documents or tool outputs overrides or hijacks the model's instructions.
What concepts are related to Prompt Injection?
Closely related concepts include jailbreak, guardrails, rag.